[01:38] my neighbor is on my wifi... is there anything I can do to scare him? [01:39] best would be to fix your wiki. [01:40] manipulating his traffic while it goes through your network may introduce legal liabilities. best to just shut it down and be done with it. [01:41] ok === duflu_ is now known as duflu === m_conley_away is now known as m_conley === m_conley is now known as m_conley_away [05:37] Good morning === maclin_ is now known as maclin === alf is now known as alf_ [09:01] good morning desktopers [09:04] morning! [09:05] morning! [09:08] hey larsu, wie gehts? [09:09] hey Laney, larsu, how are you? [09:09] happy that I work from home ;-) [09:10] weeeeeeeeeeeeeeeet [09:10] haha [09:10] otherwise good! you? [09:11] Laney: good thanks! [09:11] I'm ok, I got a cold a friday which is slightly annoying, but I'm slowly getting over it, it seems [09:14] oh I wanted to show you something ... [09:14] https://bugs.launchpad.net/ubuntu/+source/ofono/+bug/1276699/comments/2 [09:14] Launchpad bug 1276699 in ofono (Ubuntu) "scan-for-operator script fails: org.ofono.Error.Failed: Operation failed" [Undecided,Incomplete] [09:14] " we currently only support automatic network registration, and thus scanning is forbidden." [09:15] shrug, kenvandine said it used to work! [09:15] it did [09:15] ok, so the title is misleading ;-) [09:15] so do we need to take that feature out ... [09:15] but "good", it's not our fault [09:16] yeah, I'm going to talk to them though [09:16] that one seems like a basic feature, it sucks to take it out [09:16] I'm pretty sure it did work before [09:16] so maybe he is mistaken [09:16] though I've to admit in practice autoconnect is probably good enough for most users [09:16] yeah [09:17] I had to use it in the usa last time because roaming didn't work [09:17] let's talk to awe or rsalveti when they are online [09:17] k, /me patch pilots [09:17] Laney, enjoy! [09:18] larsu, don't bother arguing with that guy on the evince bug, I guess he's going to be happy enough once the keybindings issue is resolved [09:18] after the über dist-upgrade of doom [09:18] I saw all indicators landed, I wonder if that comes with any new bug [09:18] * seb128 starts update-manager [09:20] seb128: I'm not arguing, but trying to explain. I wonder why F10 doesn't open the menu on unity anymore... [09:21] larsu, because there is no cog menu on unity? [09:21] F10 used to open the global menu [09:21] I'll ask attente later today [09:21] that's an attente issue [09:21] I think he's working on it [09:21] ya, it's either an issue in the menus or the global keyboard shortcuts [09:21] both attente issues :) [09:22] man, he sure picked the easy problems on our desktop... [09:22] lol [09:22] you have to give him credits for picking those issues and sticking to them until they are fixed! [09:23] hi seb128 [09:24] darkxst, hey, how are you? [09:24] good, bit cooler now ;) [09:25] come to Europe if you like cold [09:25] or not, go to Canada I guess [09:25] darkxst, what's the plan for external panels in g-c-c-vanilla? which ones do you want to keep? [09:25] we can do wet https://witness.theguardian.com/assignment/52f0dd85e4b08e6aaede60d3 [09:26] hehe [09:26] seb128, mainly dejadup, UOA (not installed by default but we have users using it) [09:27] darkxst, UOA is going to be problematic I think, would a standalone GUI work for it? [09:27] seb128, yes [09:27] I'm saying that because robert_ancell had a look at dual building it for u-c-c and g-c-c and mostly gave up [09:27] so I think they want to do what deja-dup is doing, having a standlone interface for non Unity [09:28] gufw, but that might just be a desktop link [09:28] k [09:29] likewise with ubuntuone [09:29] darkxst: what's the parent branch of this one you gave me? [09:29] the changelog of that branch is weird [09:29] Laney, robert_ancells [09:29] I'm kind of confused [09:29] why didn't he branch off of the current g-c-c one? [09:29] I would expect a "drop " on top of the current version [09:30] he dropped everything... [09:30] and I added back in the bits that are required [09:30] that seems the wrong strategy [09:30] seb128, it wasn't my strategy! [09:30] imho you should take the current "trunk" from trusty and drop things you do'nt want [09:30] yeah [09:30] indeed [09:31] what's I'm saying is "feel free to discard what robert_ancell did" [09:31] I'd rebase on the ubuntu-desktop branch [09:31] if I were you [09:31] I doubt he has much interest in it, he just tried to get things moving [09:32] right, I don't either if we can still sneak in that gnome-desktop transition (that makes seb128 so nervous) [09:32] seb128, fwiw config migration is not an issue, since it still uses the same file, but more of a backup for persistant storage [09:33] good [09:33] well, I'm too busy to help driving that transition through, and it makes me nervous [09:33] I'm not going to block it if you find somebody wanting to take responsibility for it/do the uploads though [09:33] you can try with Laney or other [09:34] robert_ancell seems to share my opinion that we should better delay that one after the LTS, for the record (we discussed it last week) [09:35] I didn't follow the issues really [09:35] Laney, all the XRANDR code move from gnome-deskop into a mutter [09:35] basically gnome-desktop replaced their gnome_rr xrandr api to call a new dbus servic [09:35] so I ripped that code and made a standalone daemon which seems to work well [09:36] g-s-d obviously needs a bunch of git patches to adapt, g-c-c is much more trivial [09:37] gnome-screensaver and others also use that API and are going to do patching if we do that [09:37] but the bit that makes me most nervous is changing a xrandr library but a new, virtually untested for us, dbus service [09:37] seb128, its largely the same code [09:38] I sometimes thing Gnome has gone crazy w.r.t. DBUS [09:38] whats wrong with using component specific APIs like XRANDR [09:38] do a PPA with all of the packages and a call for testing on -devel? [09:38] anwyay, regarding g-c-c - can you put the changes back on top of the current branch? [09:39] lifeless, the whole point was to abstract away the display server from gnome-desktop [09:39] Laney, yes will rebase, but won't be tonight [09:39] okay, no problem [09:42] cyphermox_: do you have commit to network-manager? [09:42] upstream [09:44] Laney, if he doesn't he most likely know upstream enough to get stuff commited [09:45] either way you can bounce patches his way I think ;-) [09:45] that was my backup hope [09:45] cyphermox_: https://bugzilla.gnome.org/show_bug.cgi?id=663774 [09:45] Gnome bug 663774 in nm-applet "Country selection dialogue does not react on Enter key" [Normal,Unconfirmed] [09:47] darkxst: uhm, ok. And DBus is the right answer for that? I would have thought an abstraction library, but thats just me. [09:55] larsu, https://code.launchpad.net/~samvasko/libappindicator/libappindicator/+merge/204202 ... looks fine to you? (I just crossed it while looking at pending merge requests) [09:57] seb128: yep, looks fine to me. Shall I approve? [09:57] larsu, +1 from me, thanks [09:59] Laney, do you know what desktops use g-s-d/u-s-d? [09:59] out of Unity/gnome-shell of course [10:01] fallback too [10:01] and there are sessions that are fallback but with different window managers [10:01] Laney, what is fallback likely to use? g-s-d or u-s-d? [10:01] but you can think of those as mostly the same [10:01] asked differently, do we need to keep the legacy key grabber in u-s-d [10:02] g? [10:02] seb128, yes, I see all the fallback desktops using u- variants [10:03] darkxst, for the control-center yes, I was wondering for the settings daemon [10:03] seb128, same [10:03] I wonder if we need a rotw-s-d [10:03] rest-of-the-world-settings-daemon :p [10:03] right now, u-s-d should fit the rest of the world [10:03] darkxst, that's convenient for you, means we get to keep the legacy code there [10:04] I'm not really wanting to do that though [10:04] doesn't compiz need the keygrabber? [10:04] I wanted to take the opportunity to clean stuff we don't use/need [10:04] no [10:04] attente is moving that to compiz, same way GNOME moved it to gnome-shell [10:04] that's the only reliable way to do grabbing [10:04] right now it does though [10:05] seb128, in that case the fallbacks can you that in compiz? [10:05] e.g to key modifier only keys to work [10:05] some fallbacks session don't use compiz [10:05] although I suppose some insist on using oldschool metacity [10:05] so we can fix things for those in compîz [10:05] well [10:05] or some are Laney and use custom wms [10:05] ;-) [10:05] (can't even remember the name of that wm now) [10:06] it's still panel + gsd [10:06] xmodmap? [10:06] monad [10:06] close :p [10:07] Laney, context is https://code.launchpad.net/~attente/gnome-settings-daemon/gnome-key-grabber/+merge/202758 [10:07] well I guess it's not the end of the world to keep the legacy grabber in there [10:07] it's just more code to maintain [10:09] I see [10:09] that stuff seems pretty stable in practice though [10:10] right, as said it's just cruft/old code to keep around [10:10] the reason I want to clean it out is the same that made GNOME drop their fallback code [10:10] less code makes the job easier for us [10:10] but as said, no big deal for that one, we can keep it at least for the LTS [10:11] as it is the fallback sessions will get gsd btw [10:11] darkxst just said he expect those to use usd [10:11] seb128, I have no intentions of maintaining legacy desktops, surely that should fall onto the teams that still want to use them? [10:11] yep [10:11] but somebody has to fix it to do that [10:11] darkxst, I can copy/paste what you just wrote [10:12] darkxst, gnome-classic is not more an Unity session than it is a GNOME one, it's rather closer from a GNOME one in fact [10:12] seb128, gnome classic *IS* gnome-shell [10:12] gnome-panel-session [10:12] whatever that's called [10:13] no it is gnome-shell [10:13] with some extensions [10:13] well, what Laney is using [10:13] and that's not gnome-shell [10:13] fallback [10:13] we have users who still run gnome-panel with the wm of their choice [10:13] there are some issues with UI expectations, but gnome classic is technology wise identical [10:14] or whatever it's really called [10:14] gnome-panel aka fallback, would be much better using Unity stuff [10:14] they use the same indicators etc [10:15] edubuntu are going to switch over to Unity for XDG_CURRENT_DESKTOP [10:15] right, for UI bits it makes sense [10:15] (for gnome-flashback) [10:15] g-s-d I'm unsure about [10:15] Sweetshark: have you seen bug #1200277 is still alive? [10:15] Launchpad bug 1200277 in libreoffice (Ubuntu) "[LibreOffice] - libreoffice-writer.desktop when drag/drop to desktop, 100% broken. " [Low,Confirmed] https://launchpad.net/bugs/1200277 [10:16] seb128, the biggest problem here is there is no distinction between UI bit and backend bits [10:17] backend wise, gnome-panel etc is very similar to Unity (apart from the few extra integrations you have) [10:17] it's not that far from GNOME either [10:18] haha [10:18] you just don't want to have to keep maintaining the legacy code needed for those session (which I understand, but I don't want either ;-) [10:18] seb128, do you have any idea how much legacy code is hanging around to support Unity/panel sessions? [10:18] I think we are eventually going to need to say "if you use a non gnome-shell/compiz wm, you are on your own" [10:18] but that would be unkind to do that before the LTS [10:19] darkxst, I've been putting those patches in for the most part, so yes [10:20] right, and apart from the possible move of keygrabber into compiz, everything is needed by unity also [10:21] right, which is why we are making u-c-c/u-s-d [10:21] not sure what we are discussing at this point [10:21] out of [10:21] Laney, nobody wants to maintain legacy code for you it seems... [10:21] no one wants the legacy code! [10:22] right [10:22] well, I guess we are going to keep it in u-s-d until it's creating enough issues that we decide that those sessions to find a solution that doesn't put the burden on us [10:23] I think highvoltage does some upstream flashback maintenance :-) [10:23] need to find* [10:23] mitya as wall [10:23] well even [10:24] perhaps all the legacy fanatics can move to MATE [10:24] Laney, https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1274740/comments/21 btw [10:24] Launchpad bug 1274740 in nautilus (Ubuntu) "Re-enable traditional titlebar on 'gnome-but-not-shell' sessions too" [Low,Confirmed] [10:24] Laney, we were discussing that last week [10:24] altough I have no idea where that is at [10:24] yeah, me neither* [10:24] seb128: nice, I can't remember the details from before [10:25] Laney, it's early enough in the cycle than we can deal with issues [10:25] I made some patches over the weekend to set that for the indicators ;-) [10:25] and I trust Dmitry to know what he is doing [10:25] nice ;-) [10:25] Laney, Jeremy switched over flashback to Unity in past, then straight away reverted it since it was buggy [10:26] do you remember what was buggy? [10:26] no, but I checked recently, as far as I can tell that was only due to the OnlyShowIn fields in the various desktop files [10:27] gnome-panel, nautilus-classic and indicator* [10:27] ok, that's what we were discussing, Laney remembered that it was tried but not the reasons it didn't work/what was to be resolved [10:28] indicator ones will need a autostart condition I suppose [10:28] GNOME3 unless-session Unity [10:28] we use upstart for those in Unity [10:28] oh, then it should be fine [10:28] so they can do what they need with the desktop [10:28] which is just add Unity to them [10:30] offtopic, we have been hitting a weird bug where gdm restarts on upgrade, could that be some upstart thing? [10:30] wasn't that https://launchpad.net/ubuntu/+source/gdm/3.10.0.1-0ubuntu2 ? [10:31] apparently not, that did work at the time when I tested it, but its still happening now [10:31] but I doubt it's an upstart thing [10:31] is gdm upstart managed nowadays? [10:32] if it is the job didn't change in ages and usually they don't do restart, just start/stop and the stop for a dm is on init level change [10:32] I wouldnt have thought so [10:32] however, when I built and test the 3.10 package it was fine [10:32] (i.e Dec sometime) [10:33] can you reproduce the issue? [10:33] sure, just reinstall gdm while gdm is running, and it brings down the entire session [10:35] what does "sudo status gdm" says? [10:36] does it restart when that happens or just goes down? [10:36] seb128, it restarts, but takes some time [10:37] hum, k [10:37] I can try having a look at it in a vm today [10:37] I can't even switch to a vm once it goes down [10:38] that seems like a kernel/xorg issue to me [10:38] but ssh says: gdm start/running, process 4064 [10:39] and apparently this time it is coming back up [10:39] no segfault/sigabrt in the Xorg.* [10:39] no apport repport? [10:39] did the pid change? [10:39] no crash logs [10:40] X still seems happy enough [10:40] no idea if the PID changed [10:40] you can maybe try to sudo gdb -p $(pidof gdm) [10:41] (gdb) c [10:41] then reinstall [10:41] and see what happens in the gdb [10:41] do that from a ssh [10:42] seb128, x tried to restart and failed [10:42] why did it try to restart? [10:43] debian [10:43] I don't get any logs of that [10:43] gdm tries to fire up on a new dispaly (:1) and fails [10:43] when did that issue start? [10:44] but on my real machine, upgrade will result in gdm restarting [10:44] late Jan [10:44] k, so not the new xorg [10:45] I've problem with user switching for some days, I blame new xorg/intel [10:45] but that's probably different [10:51] seb128, sometime before 22nd Jan [10:52] hum, k [10:53] and my gdm changelog is dated way back in Nov [10:54] but it was certainly tested through dec [10:54] what do you mean? you don't have the recent fix from trusty? [10:54] https://launchpad.net/ubuntu/+source/gdm/3.10.0.1-0ubuntu1 [10:55] Of I have the recent fix [10:55] ^of course [10:56] k [10:56] I can try having a look later, need to set up a vm or try on my real box (but testing needs closing/restarting sessions) [10:57] seb128, it will happen in a vm too [10:58] good, makes debugging easier ;-) [10:58] just gdm fails to actually restart in my vmware VM's, whereas it does on real hardware [10:59] * darkxst could not live without VM's although I am running out of diskspace [11:02] darkxst, https://bugs.launchpad.net/ubuntu/+source/gdm/+bug/1266496 seems still valid (look through recent reports) [11:02] Launchpad bug 1266496 in gdm (Ubuntu) "extra "fi" in file" [Undecided,New] [11:05] ooh, just started snowing [11:07] Laney, that's another kind of weet ;-) [11:07] an exciting one [11:07] yeah [11:07] but it won't settle because of the previous wetness [11:10] Laney, can you bring some snow ;) [11:11] sure, I'll pack it inside a thermos and send it over [11:12] Laney, a 40ft container would be better ;) [11:16] seb128, right, though I can't see how that would cause this issue [11:16] but will fix [11:17] yeah, it's not likely it, I just crossed the bug while looking at reports [11:17] thanks [11:22] seb128, I havent seen a single report of this, since people only submit bugs when prompted right? and gdm doesnt update that often... [11:22] right [11:22] woah, what happened [11:22] 11/02 11:20:15 seb128: can you push e-d-s please? [11:22] Laney, did your box reboot by itself again? [11:22] did that send? [11:22] freenode lagged out [11:22] Laney, "push"? there is a vcs? [11:23] this is on a different server :-) [11:23] ~ubuntu-desktop/e-d-s/ubuntu [11:23] oh ok [11:23] sure, sorry about that [11:23] cheers, np [11:35] Laney, ok, vcs updated [11:35] merci! [11:35] de rien [11:43] mlankhorst: was that ld crashing bug only on i386? [11:43] i.e. not amd64? [11:43] yeah [11:44] mmk === MacSlow is now known as MacSlow|lunch [12:41] MUAHAHA =) http://people.canonical.com/~xnox/gtk3-qmlscene.png [12:41] this is qmlscene, loading Desktop Components gallery.... using gtk3 instead of gtk2 =) here be dragons. [12:58] xnox, dude ! [12:59] and it's only a 260 lines patch so far. https://codereview.qt-project.org/#change,77853 [13:05] xnox, I'm not sure to grasp what that is doing [13:05] xnox, you are making qt use gtk widgets? ;-) [13:05] yeah [13:05] evil ... [13:06] that will get you all sorts of political issues :) [13:06] seb128: more specifically, it already knows how to use gtk _2_ widgets, i'm making it use gtk _3_ widgets/themes instead =) [13:06] that seems a weird concept [13:06] ogra_: my evil plan is to rewrite ubiquity using qt/qml and nobody to notice the difference. [13:06] haha [13:06] seb128: well, do you find mumble & skype out of place in terms of theming on ubuntu? [13:07] xnox, they do, but I would think the fix there is to make a Qt theme matching our desktop one [13:07] (e.g. the buttons are fully pale orange, instead of having gtk3 orange aura) [13:08] seb128: Making qt themes is a pain, and at the moment, it simply dlopens gtk2 renders widgets and copies the pixmap back into QImage and slaps it onto QWindow. [13:08] seb128: and that ends up looking far more native, then trying to maintain a parallel theme. [13:08] sans gtk3 and gtk2 theme missmatches. [13:08] k [13:09] firefox is moving to gtk3 finally (patches posted), so it's time for qt as well ;-) [13:09] then we will be left with google chom[e|ium] and libreoffice. [13:09] eclipse / swt did move to gtk3, so i'm not sure what's holding libreoffice back. [13:10] hrm, won't switching firefox to gtk3 break flash? [13:10] or is that handled by the firefox sandbox now? [13:10] mdeslaur: no idea, but as far as i remember firefox is sandoxing flash. [13:10] mdeslaur: but i also thought flash stopped publishing napi plugin, and it's pepper only as part of google chrome.... [13:11] chromium is dropping gtk and moving to their own widget set in the latest dev version I believe [13:11] xnox: well, the old flash is still supported for another 3 years or so [13:11] excellent =) or i guess i'll make my mind up once it lands in stable channel. [13:11] it looks like ass at the moment :P [13:12] well, like windows ass :) [13:12] mdeslaur: based on my timelines we will be all using silverlight by then.... oh wait maybe not =) and it will be more like XP never-ending EOL. [13:12] (by ass, I mean out of place) [13:12] xnox: lol :) [13:12] mdeslaur: http://www.newyorker.com/online/blogs/borowitzreport/2014/02/gates-spends-entire-first-day-back-in-office-trying-to-install-windows-81.html [13:13] hehe, yes, saw that...funny :) [13:13] it took me 2 days to figure out how to upgrade stock OEM 8 to 8.1 [13:13] I booted windows 8 on my new laptop just out of curiosity before I blew it away. I had to google search how to turn it off. [13:13] seriously. [13:14] turns out pre-loaded crapware, well McAffe antivirus, blogs connections to the 8.1 update servers and thus download just stalls at 0% and no indication how to fix it. [13:14] lol =) i push power buttons to turn it off =) [13:14] heh [13:21] xnox, so, doing an apt-get install --reinstall gdm leads to a console output [13:22] Preparing to unpack ...deb [13:22] gdm stop/waiting [13:22] Unpacking gdm [13:22] Setting up gdm [13:22] gdm start/running [13:22] xnox, do you know what could make gdm change to stop before the unpacking? [13:23] the upstart script has [13:23] stop on runlevel [016] [13:23] that doesn't make sense to me [13:24] seb128: dh_installinit by default stops in prerm and start in postinst === MacSlow|lunch is now known as MacSlow [13:24] seb128: one should pass --no-scripts or something like that, to make sure restart is not performed. [13:25] seb128: it's not events that trigger restart, it's update-rc.d in maintainer scripts that explicetely stops and starts it, just like any other daemon. [13:25] mpt: heh, have you seen this? http://www.therestartpage.com/ [13:25] xnox, thanks [13:26] Ahaha [13:26] mdeslaur, that’s brilliant [13:26] it's even animated! [13:26] love the bios screens [13:27] got surprised by the mac startup sound :) [13:28] 2gs is quiet. :P [13:28] mdeslaur, reminds me of http://www.clivejames.com/poetry/james/windows [13:29] lol [13:30] mpt: kind of like this: http://slowcoustic.com/wp-content/uploads/2010/12/The-Birds-is-coming.jpg [13:30] hehe [13:30] did I miss the os/2 one? [13:31] ah nm found it [13:43] seb128: I know scanning was working fine with ofono in the past, as I implemented that feature [13:43] not sure if it was disabled later on by awe [13:43] will check once he's back [13:44] rsalveti, ok, no hurry ... is he away this week? when is he back? [13:44] Laney, ^ === alex__ is now known as alex-abreu [13:45] seb128: he was sick yesterday [13:46] k [13:48] rsalveti: thought as much! [13:48] rsalveti: can you confirm if scan-for-operators / manual mode in system-settings works for you? [13:48] and it's useful, I use it all the time when traveling [13:49] sure === greyback is now known as greyback|lunch [14:07]